Basic questions we should ask about social skills before theorizing about autism

It seems to me that research on social ability and cognition, particularly in autism, has skipped over some crucial questions.  We're trying to understand what goes wrong in autism without understanding, even among neurotypicals, what social success is, what contributes to it, or even whether it exists in any sort of universal way.

1) Is there anything universally human about social success?  
a. Do socially successful people have any characteristics across cultural and subcultural:
  • Beliefs about the mind, psychology, emotions, etc.
  • Norms about the goals for social interactions
  • Norms about appropriate display rules for emotions, body language, etc.
If not, we should be wary about any claims of an evolved "social module" in the brain, claims that people with autism lack such a module, and studies that compare autism prevalence rates across cultures.  
b. Do they have any common behaviors across the same?  
Even if the personal characteristics of socially successful people are the same across all cultures, they might express these through very different behaviors depending on the culture.  In diagnostic research, this may affect the characteristics by which we recognize developmental delays.  For clinical researchers trying to develop social skills programs, it determines what behaviors to teach.

2) How much does the average neurotypical adult actually understand about other people's thoughts, feelings, and motivations?  How much better are they than chance?  What do they understand that an autistic person does not?

We've assumed that they have extremely accurate "mind reading" ability, but everyday experience and certain lab experiments would suggest that we are quite inaccurate, sometimes approaching chance.  We assume that neurotypicals have knowledge about other people's mental states that others do not have, but do they?  Maybe they're as clueless as autistic people, but they have better procedural abilities--e.g., automatic facial expressions and gestures, approach/withdrawal behaviors, mirroring others' actions, and timing.  If this is true, we should be wary both of exaggerated claims about our own "mind reading" capabilities and autistic peoples' social cognition deficits.

3) We have fairly sophisticated concepts about our own and other peoples' minds, but how much do these actually influence our real-world social behavior?  Does it matter whether these concepts are implicit or explicit?  Do people with more elaborate explicit theories of other people's minds have better, worse, or no difference in their social success?

If explicit "theory of mind" alone produced social success, most people with Aspergers would be social geniuses, and many NTs would be social dunces.  But does it help at all?  In my completely anecdotal experience, high intelligence tends to lead to more elaborate explicit theories about others' minds, but lower social success (except around others with similarly high intelligence).  This may be false, but implies the relationship between theories of mind and social success may be more complex than is often supposed.

4) What perceptual abilities are necessary for social success?  What perceptual abilities are helpful, but not necessary, for social success?

Most of us realize that reading requires us to do a lot of sophisticated things: perceive and recognize symbols in the appropriate order, move our eyes in a coordinated way across the page, call up the appropriate phonological associations (e.g. what the letter "b" sounds like), remember words and their rich associations, and build these already very complex word representations into sentences, ideas, and narrative.  When we read, we are at least can control how fast we read, and can stop and start at will.  Social interactions are likely to be even more complex because we lack even this control.  What visual, auditory, somatosensory, or interoceptive processes does it take to observe, integrate, and rapidly respond to other peoples' facial expressions, gestures, and other movements?  If someone has a deficit in one or more of these, can they make up for it and still achieve social success, and if so, how?

5) What perceptual, cognitive, motivational, emotional and other information do we integrate when deciding how to act during a real-time social interaction?  How much information do we have to integrate, and how much occurs on any sort of conscious level?

6) What relationship does emotional empathy--the ability to feel another person's emotions or pain--have to social success?  Does it help, hurt, have no systematic effect, or have some more complex relationship that depends on other factors?

Many people assume that the more emotional empathy a person has, the more social success they will have, but this assumption should be tested.  My guess is that emotional empathy interacts with performance in a similar way to anxiety: too little and too much are harmful; just the right amount helps.  Too little and you don't care sufficiently about others to treat them well; too much and you are too overwhelmed to think of anything other than making the feeling go away, and may not behave appropriately.  Until we understand this relationship, we should be wary of claims that autistic people lack emotional empathy, or that lack of emotional empathy causes their social failures.

6) For people who are perceived as belonging to an outgroup, or just different, are there any ways to overcome this and achieve social success?  If so, what strategies work?  (Note that the "outgroup" includes not only obvious gender, ethnicity, class and religious differences, but also personality and neurodiversity differences, such as extreme introversion, a developmental disability, extremely low or high IQ).

If an autistic person or an extremely high-IQ person can achieve social success by emulating some neurotypical behaviors, then we should teach them these skills and they should learn them.  If an introvert can only fit in by becoming an extrovert, an autistic person can only fit in by becoming completely neurotypical, and a high-IQ person can only fit in by disguising all traces of their true thought processes, then we should stop blaming their social failures on lack of social skills and instead focus from an early age on placing them with others like themselves so they can find true peers.

7) How aware are we of the processes underlying our own social interactions?
In my own life, the social advice I have received from extremely socially skilled people has been uniformly unhelpful and simplistic (e.g., "just be yourself").  Are socially successful people like other experts, in that they just do what they do without being able to explain how they do it?  To the extent that researchers on social success are socially successful themselves, how much does this distort their research?  How can we work to make the implicit visible?  Ideally, how much social success should a researcher have? ;) (Too little, and they will have trouble grasping the phenomenon well enough to study it; too much, and they will have overly simplistic ideas of how we actually accomplish it).

8) We know that loving, emotionally intelligent parents can foster social success in neurotypicals and that cruel, emotionally abusive parents can stunt it.  Yet, in order to avoid the refrigerator myth, we assume that parenting has no effect on people with autism.  Is this a justified assumption?  How does the family and educational environment interact with an autistic person's genetic predispositions?  What is the best case outcome for an autistic person?

I don't want to unfairly malign an entire field, and I'm always interested in learning about interesting research.  So, do you know of any researchers already trying to address these questions?  If you were a researcher, how would you go about answering them?